What are a hospital's costs? | Hacker News Reader

What are a hospital's costs?

nytimes.com

56 pointszabramow77 comments

The group began with head and neck cancer, treatment of which turned out to involve 160 processes requiring measurement. To assess outcomes, it asked patients which they thought were most important. Head and neck cancer patients wanted to be able to talk and to swallow. (Survival, which many doctors had thought was a top priority, was not something patients raised; many assumed they would survive.)

That parenthetical note is perhaps the clearest illustration I've ever seen in my life on the communication and expectation gap between experts and non-experts.

I was looking in the article for a followup to that statement, but there was none. The reason I raise an eyebrow is simply that even in the absence of a communication gap, talking+swallowing as top concerns seems like a reasonable response.

What matters more to you, being alive, or having a tolerable quality of life? Are we sure that there's a misunderstanding that death is a likely outcome (I could believe it; but I would also be slightly surprised, since "cancer==death sooner or later" exists even as a trope in mainstream media) as opposed to just a prioritization over things that matter so long as you're alive?

Anyway, this was quite a tangent from the original article content, just something I was mulling over given your comment.

> Are we sure that there's a misunderstanding that death is a likely outcome

There isn't a misunderstanding that death is a likely outcome, there commonly is a misunderstanding of survival time. Gawande's book covers the subject of differing expectation between doctors and patients when discussing harsh treatment, patients think it'll give them months or years, doctors think weeks.

If that isn't explicitly put on the table — and it usually isn't — you can get drawn-out "medical fights" where the patient's quality of life is more or less negative.

Atul Gawande's "Being Mortal" has a chapter or two on expectation gap, the realisation of how large it was made my stomach drop.

And do so once more when Gawande described how the expectation gap still existed when doctors become patient (both he and his father are doctors, after his father fell with cancer he first succumbed to the expectation gap as he was preparing a book on the subject).

It's a great, eye-opening book.

edit: http://www.pbs.org/wgbh/pages/frontline/health-science-techn... covers some of it

In my experience that is to some extent a logical assumption though. Most doctors will "fix" you to the best of their ability, it's the life quality issues you really have to fight for.

It is why my wife did not choose the first two surgeons. For her being able to smile was most important. Then to see was the second. For both of those she wanted to have into old age. Losing hearing and balance immediately were of no concern. She just assumed a total resection and good recovery.

I was in the ER with what I expected was (and turned out to be kidney stones). I was stunned when the doctor's first question was something like "what is your expected outcome of this visit?".

I mumbled something like "make the pain go away and not come back".

I work in the healthcare industry and have a good friend who works at a major hospital. You know what I've learned?

Hospitals have no clue what it costs them to serve an individual patient.

The lowest level where they understand costs is at the service level (e.g. ER, cardiac lab, stroke center). It's one of the reasons why you see all the wonky prices from hospitals: they just "create" prices for individual procedures until it covers their aggregate costs. It doesn't really matter to the hospital if procedure A is $1K and procedure B is $5K. As long as Arate + Brate > costs, they are fine with it.

Well, yes, that's what the article says, but highlights one hospital working to change that.

There is a clinic in Houston that was profiled earlier that did price per procedure and knew their costs. They were not a Medicaid / Medicare provider which might have made a difference.

The idea of pricing out hospital procedures as if they're widgets or oil changes is ridiculous to begin with.

I don't know that anyone is suggesting that.

I'd also argue a routine diagnostic procedure, like a blood test, is pretty comparable to a mechanic doing an oil change.

Is calculating statistical aggregates over a long period of time ridiculous? Many service based procedures have uncertain length, flipping a coin to arrive at a price doesn't seem like a great solution either.

Depends on the procedure doesn't it? You might not price open-heart surgery that way since you want a great doctor, price be damned.

But what if it's not so serious, and not an emergency? People have rotator cuff issues that they take years to get fixed regularly. Or a hip replacement? That's rarely an emergency but rather a solution to long term pain that's not painkillers.

I think the other thing a lot of people would like to see is some kind of doctor and hospital report card so that you could compare on price AND quality. Which would be great as it'd force the terrible surgeons who regularly do such a bad job (accidentally or on purpose) that their patients have complications, die, etc out of that line of work.

Obviously there would have to be some kind of risk classification system so that doctors and hospitals that take on high risk patients wouldn't be punished for doing so, but I think that some kind of "hot or not" style risk assessment system could be developed whereby doctors spend 15 minutes a day reviewing risks anonymously for other doctors and each case gets rated 5 times and that determines the risk score to hopefully remove all bias.

I think everyone misunderstood my point, which is that apportioning those costs on a procedure level is a fiction. Almost all of your costs are fixed: salaries, facilities, equipment. And the amount of procedures you can do with those fixed resources varies dramatically.

It's definitely worth quantifying the cost of providing medical care, but doing it in a more fine grained way than at the department level.

If I may give some feedback: I don't know how anyone could have read your post as being about the appropriate level of economization of medical care due to the relative importance of fixed vs variable costs. Your comment says nothing about that, and just comes off like the usual screed about "how dare you apply accounting to health care, this is too important for that!"

I think you could have been more explicit about (what you now say was) your core point.

> Almost all of your costs are fixed: salaries, facilities, equipment. And the amount of procedures you can do with those fixed resources varies dramatically.

Yes it does vary dramatically, but there's no way there's 3 orders of magnitude of variation and definitely not 2 orders. There might be a single order from top to bottom, a 2 hour surgery versus a 20 hour surgery or a 15 minute visit with a doctor versus a couple of hours with a doctor.

Lawyers have the same problem and while they do bill by the hour, they can generally give you a pretty good idea of what the more basic things might cost to handle. I'm not suggesting that they could give you a +-3% estimate for your SCO vs IBM lawsuit, but if it's a DUI or a will or a trust or whatever, once they've done it a few times they know how the game is played.

I totally get that people aren't courts and doctors aren't lawyers, but I think it's a decent counter-example to the notion of "it's hard so why bother trying?"

Care to elaborate? In medicine, just like almost everything else, you will find a "standard" for each type of procedure. Do they sometimes take more or less time? Yep, but that all plays into the standard over time. So when you have an outlier of the standard you can evaluate what went wrong or what was different to better understand the variables that play into these elements of variance. Medicine is a science. The more we treat it that way the better... you cannot just throw your hands up and deem it ridiculous.

Why not?

Flat rate mechanics have a book that says repair X takes Y many hours. If it takes the mechanic less, the shop benefits. If it takes more, the shop loses. But in the end, only Y many hours are billed.

Why can't it work similar for hospitals?

I'd guess the OP's argument would be that if you come in with some symptom finding the underlying cause and treating it is a highly variable process. True, but not really so different than taking your car to the mechanic and telling them it makes a funny noise.

Particularly since in so many cases, you only go to the doctor to get the signature for the appropriate drugs, or to get the specialist work done, for a problem that is already known. That's not a whole lot different than going to the mechanic and telling them to change the spark plugs or flush the transmission.

Forgive me if I'm mistaken, but isn't a "procedure" something that is well documented set of steps to be taken to remedy a person's ills or assess what is actually wrong with the patient (ie. mri, xray, etc) ?

Sure, like with anything, there may be some unknowns that cause some variations in the implementation of the procedure, but it's likely a valid (in most cases) way to track performance and control costs. If some hospital is regularly charging 2 standard deviations more than other hospitals for same procedures, that seems it would be a red flag.

Or if a doctor is regularly taking 2 standard deviations longer to perform same procedures as the rest of his/her peers, that's a red flag.

> if a doctor is regularly taking 2 standard deviations longer to perform same procedures as the rest of his/her peers, that's a red flag

FYI defining the "peers" part of that calculus is the hard part. Physicians choose their patients. Those that choose low hanging fruit do quite well in metrics. It doesn't require much skill to pick low hanging fruit. It turns out that those who are good at picking fruit from the top of the tree, tend to be called on to pick fruit from the top of the tree.

Individual procedure codes in no way communicate anything about difficulty.

I hear you for sure. I think similar concepts apply in most non-trivial pursuits. I'm certain I can't generalize across the board, but typically my assumption would be that the person who is "better" or more experienced with certain things are getting the more difficult cases, and so perhaps by either separating them with a "senior" or "specialist" title and different pay scale they're not being graded on the same curve as people who are handling the more routine cases.

Ultimately I think those edge cases can be fleshed out and monitored with regular auditing.

> but typically my assumption would be that the person who is "better" or more experienced with certain things are getting the more difficult cases,

You sure about that? I also use my experience to filter out clients that are going to be a huge pain in my tail. When I was less experienced I ended up with some jobs that were huge messes because I didn't see the warning signs. Now that I can better predetermine the work involved I can choose not to take the job, or I can correctly tell the client the job will be much harder than it actually appears to be.

I think we're talking about different things. I too do my best to weed out potentially troublesome clients. But when a larger group of people have a "body of work" they're responsible for, that's the sort of hierarchy I would expect.

In the long run, though, I think (in an open market system) always turning down hard cases (for example referrals from peers, or people who specifically seek you out for your expertise) can negatively affect your reputation. In the end I might call it a wash because the person who is truly trying to do the least amount of work for the most personal benefit wouldn't try to stand out by finishing their work substantially faster than their peers or charging substantially less. So you probably lose a lot of the marketing benefit of out performing your peers since your ideal scenario will be to remain hidden in the pack. And by doing, it seems reasonable to assume people won't be trying to knock down your door to get you to do work for them like someone who is accepting the more challenging work and doing so successfully.

Free markets are many things, but they are never meritocracies without the assumption that capital is merit. Whoever has the most money always has the best marketing.

I may be more cynical than the average person on this, but I have to believe, for the sake of human kind, :) that even the casual observer today doesn't believe capital is merit.

At least not as a broad generalization. However of course people who are good at what they do generally have more opportunities to make more money than those who are not.

On one hand, you're right, but on another there do seem to be some "objective" measures of quality too -- for instance, hard patients or not, sowing up a patient with a sponge left in them is a serious error (that is also not all that uncommon, unfortunately).

Quite often, these sorts of medical errors are reduced significantly by adopting risk and personnel management practices that were learned from a dark chapter of air traffic control. These practices don't mesh well with the current procedures performed compensation structure which is one of the reasons that developing new reimbursement models is en vogue. The status quo has enough inertia that it takes a major mishap to get each institution's bureaucracy to react.

> The idea of pricing out hospital procedures as if they're widgets or oil changes is ridiculous to begin with.

That seems pretty silly. There are plenty of procedures with marginal effect on survival and/or quality-of-life, and a health-care consumer will have varying preferences for them.

A rich man might be willing to spend an hour in an MRI if he's sprained his ankle; a poor man might prefer to keep his money and walk it off. Neither is wrong: they just have different preferences.

For that matter, given a choice between a $15 pill of Tylenol and dealing with an ache, I'd take the ache every day of the wee.k

First off, why?

How on earth could you possibly makes a cost vs. value trade off if you don't know the cost? You might have a hospital with better outcomes that charges 1/2 of that of another hospital with worse outcomes. Unless you know the cost, you'll never make the system efficient.

Medicare already does it. There are ~700 diagnosis codes that capture everything you might need to do to a patient admitted to the hospital. The diagnosis codes are linked to a dollar amount that is supposed to cover everything (except labor which is a separate payment).

This is absolutely true. But the question is why are they stuck at that rudimentary level?

My wife is at the center of this. She recently left a job as "Director of Budget and Reimbursement", and now works as "Directory for Regulatory Affairs" for a healthcare financial consultancy. So she's been the one directly responsible for the way the hospital accounts for its costs (or fails to do so).

The primary problem, at least from her perspective, is that they're effectively forced to do it this way. Medicare, and Medicaid which largely rides on many of -care's definitional coattails, is a monopsony player in the market. When you hear people claim that Medicare is more efficient than conventional commercial insurance, that's significantly because they offload much of their administrative burden onto the healthcare providers. Every hospital has a department of several FTEs whose job it is to track a defined set of metrics, including various cost parameters, patient demographics, etc., and report them to the government (the so-called "Medicare Cost Report" that my wife and her department spent a month or so on every year). That is, getting paid by the source of the largest part of the hospital's revenue entails tracking and reporting the way Medicare wants you to do it.

So that right there forces hospitals into watching their costs according to a scheme effectively mandated by the government. Thus, they're not tracking the data in a way that would give relevant cost accounting data in the way you expect it.

There's your proximate cause. But it still begs the question, "why don't they also track costs in a rational fashion, actually reflecting reality?". Part of the answer is that their margins are so low that they simply can't afford a second set of accountants and the procedural overhead that it would require.

That's not a wholly satisfying answer, because hospitals are the only industry I'm aware of that still makes extensive use of secretaries. You'd think that if they need to be so conscious of costs, they'd be looking at the kind of personnel expenses that most of private industry has figured out how to streamline. Conversation with my wife hasn't yielded any insight beyond that.

>> As long as Arate + Brate > costs, they are fine with it.

I've been advocating that we require them to charge the same price for a given service to all patients regardless of insurance or anything else. To clarify, we should not dictate the price, just that one provider charge the same price to all its patients. I never would have thought they couldn't even figure out what those prices need to be.

Any business that doesn't even have a clue about these things is clearly rolling in dough and probably corruption.

I've had the same thought. It'd add clarity (as in, an actual answer to "WTF will this cost me, after insurance?"), reduce complexity ("well, we have different rates depending not only on which insurance provider you have but on the specific plan you have", and associated personnel to manage that and similar BS) and, most importantly, it'd incentivize insurance providers to work collectively to drive down prices—I suspect they'd avoid free-riding by forming associations for which a given hospital accepts all or none of the member insurances.

Or we could just have single payer. But if we can't have that, I think the above would be a fair bit better than the mess we have now.

That's great for marginal costs. What about fixed costs?

The lobby needs to be mopped. The parking lot needs light. The lawn probably should be kept from turning into a forest. The heat should run in winter, and the cooing in summer.

Which hospital users get stuck paying for those things? What is your fair share?

You're missing the point here. The idea is that for you to have an a chest x-ray done, or an appendectomy, or a hip-replacement, is priced at the same rate as if I have the same work done. That they're not charging my insurance company X, while charging your insurance company Y, and Medicare Z, for the same service.

Another cause of this is that services are generally not priced on a case by case basis. Hospital inpatient claims are priced based off of DRGs (Diagnostic-related group). An insurance company is not charged on a line-by-line basis. Rather the entire claim is submitted to the insurance company, who will process the claim and assign a DRG to that claim based off of the procedure codes, diagnoses and other factors. It might also assign a severity level. Based off of the DRG and severity level and the negotiated fee schedule a price will be assigned. This price might be a per day amount or a price to cover the entire span.

So you have an added layer of abstraction between what the service is costing and what the hospital is being paid. This system was put in place to simplify the pricing procedure so that it did not have to be done on a case by case basis. But it limits most hospitals to only really caring to this higher level of detail.

After working in healthcare myself for a while, I've always wanted to see it be a service that was paid for by the customer directly (at least hypothetically). This concept of not knowing costs would be completely unacceptable in any other industry, but it seems that people are reluctant to price things partly out of a fear of being able to do cost-benefit analysis on care and thus on human life. Direct billing would force hospitals and other care givers to be able to give cost estimates up front, and then market forces would help bring costs down as providers would compete and patients could shop around. I'm aware that this might not be optimal in all situations (eg emergency care), but runaway costs and opaque cost-benefit relationships are not helping things.

I remember hearing that 20% of total healthcare costs are spent in the last 2 weeks of people's lives, and perhaps having more transparent costs would help avoid these costly, frequently unpleasant, and ineffective interventions.

Of course, transitioning to this would be impossible overnight, but it's an interesting thought experiment. There would still be the option for catastrophic insurance, but otherwise it would be interesting to see market forces play a role.

The problem with asking patients to shop based on price is that the penalty for making the wrong choice is much more devastating than with just about any other good or service. I did a ton of research when I bought my car, but even then the price of making a sub-optimal decision was limited to a substandard experience or waste of money. When choosing a surgeon, however, the cost of a wrong decision will be perceived to be much higher—shoddy work there could be life-altering/ending.

Of course, shoddy workmanship on my car could also have devastating results (and safety was a big part of my decision process, especially since my father might not have died in his car accident if he had side curtain airbags), but it's not considered with the same level of urgency.

Also, I'm extremely wary of pricing schemes that disincentivize preventative care and early interventions. Many illnesses are much more treatable—at much lower cost—if they're caught early. While we also need to be careful to avoid unnecessary expense, we don't want to discourage people from seeking care at the early signs of trouble.

That being said, I do agree that transparency in healthcare costs is very much needed. However, we need to be careful in considering how market forces impact personal healthcare decisions.

You make some good points regarding quality, but there are a great many things where that doesn't matter. Yes if it's a major medical issue where a life hangs in the balance quality matters tremendously. But a huge range of medical care is about quality of life or other more minor things.

I also think that if prices were available to people they might choose the preventative care themselves versus waiting until things are so bad that a major and costly intervention is needed.

People talk about externalities as if they're the devil when it comes to climate change, but when it comes to medicine nobody bats an eye. Right now the consequences of people smoking, eating poorly, drinking too much, etc are generally externalities for individuals because either their insurance eats the cost or the public does. I'm not suggesting that people should be forced to bear that cost entirely on their own without insurance, but with no pricing available GPs have one less tool to "scare straight" their patients.

I think what's especially galling is that the hospitals do have all the pricing information available in a database that they use to turn patients charts into bills. It's just not published publicly so patients can't make a slightly more informed decision on the front-end, they only see it AFTER care has been provided.

> I also think that if prices were available to people they might choose the preventative care themselves versus waiting until things are so bad that a major and costly intervention is needed.

Insert analogy with car maintenance here. I'm just not seeing it in practice given the car maint example, although it sounds nice. Clearly, based on the car example, you can't motivate people merely with $$$ or inconvenience or some responsibility guilt trip.

I'm thinking pain might work. "The total pain experience of a root canal is Y% vs flossing at a total pain experience of Z% make your choice..." Then again people seem impervious to pain, or impervious to learning from pain. "lets have another kid" "lets go on a fifteen mile hike in brand new boots" "hold my beer and watch this" If $, inconvenience, and guilt trips don't work, and pain also fails, I'm not sure whats left as a possible motivator.

Pain, cost, inconvenience, etc are all marginal motivators even if they don't have the 100% effect on people's behavior that you'd like. Some people are going to be dumb full stop. But for everyone else, giving them more information is probably going to be useful and on the whole, will move things in the right direction.

Keep in mind that I said "might" not "definitely will for sure 100%" so that's in line with my above comments re on the margin.

I think our disagreement might fundamentally revolve around you seeing "might" and I'm observing "almost always does not".

I'm still interested in the idea of listing and evaluating motivators, so we've listed four of varying opinions of effectiveness, it would be interesting to identify a fifth (or more) of equal or better effectiveness.

Thinking of what PR people like to use, there's sex appeal, I suppose that could be used in public service TV commercials as a new fifth strategy. I have no idea how to sell "preventative EKGs are sexy", but I'm no marketing guy, and somehow they sell cars and junk food with sex appeal, so its probably possible to sell EKGs and immunizations and diabetic glucose test screens under the same strategy... somehow. This sounds like such a bad idea, I hesitate to declare it a serious fifth strategy idea.

Fair enough. I guess part of the appeal of having published prices is what other people could then do with them, too. Not just that I could shop around for rotator cuff surgery or a hip replacement, but actually interesting stuff that's only done internally right now.

If prices were published then someone could track the prices of a great many procedures and look for procedures that were getting cheaper. You could then interview people at institutions that are able to do those procedures less expensively and publish it. Other institutions might then follow suit and the savings could be shared widely. The lessons might cross boundaries and spark other folks to recognize that X is very similar to Y and they can use the same lessons to make Y cheaper, easier, more foolproof, etc.

Further, with reasonable pricing information and some measure of quality it would be pretty easy for people (and perhaps insurance companies) to encourage medical tourism of a great many varieties. Maybe the country doctor or dentist is very skilled but wanted the country lifestyle and doesn't want to drive 2 hours a day to the city. But perhaps city folks would be happy to drive to the country to enjoy a cheaper procedure, hospital stay, etc versus in the city. Right now this kind of arbitrage isn't possible again because prices aren't disseminated widely.

Finally when we're talking about human lives which we think are intrinsically valuable and arguably priceless, then I'm not sure that "almost always does not" doesn't mean that something isn't valuable. Young people with families buy life insurance against something which "almost always does not" happen, namely their premature death and yet nobody argues that life insurance shouldn't be allowed. Similarly for any one individual wearing a seatbelt "almost always does not" do them any good since car accident rates are very low, at least on a percentage basis. http://www.forbes.com/sites/moneybuilder/2011/07/27/how-many...

There's a balance to be struck, for sure, but cost transparency as an alternative to complete and total cost opacity is a move in the right direction.

My daughter fell at lacrosse practice and hurt her wrist. We had an x-ray, which proved inconclusive with the swelling, another x-ray after the swelling subsided, which also resulted in an inconclusive result, followed by an MRI which indicated a bad sprain, for which we ended up getting a very cast-like splint.

This is basically the same as what we thought from the initial x-ray, as even then the doctor was advocating for a splint in the absence of an obvious break.

If costs were more apparent, we probably would have had an x-ray once, and then gone with the splint, but price opacity and insurance obfuscated the costs enough that we naively opted for the more diagnostic approach, which proved to cost a lot more than just having gone with the splint initially. As a developer, I'm inclined towards the "know what you're fixing before building patches" approach, so the diagnostic approach was logical to me, but had I known that the end result was basically the same either way, I could have saved myself the cost of an x-ray and MRI and just skipped to debugging.

If you turned the tables on your "free market" anecdote, what is the market value of that ER doctor or radiologist as a software developer? I'm not even sure how to account for emotions, I'm pretty dispassionate about unit testing frameworks, but if my daughter might be crippled for life I'm fairly certain I wouldn't be a rational market participant. Its highly likely that your trying to practice medicine without a license holds the same market value as a doc trying to be a software dev, that being somewhere between 0 and a modest negative number.

You could extend the analogy by taking printouts of the pricing plans for github(tm) and visual source safe(tm) and waving them in front of the er doc to ensure the er doc makes the wisest possible free market decision when selecting a source code repository. This is only a fair analogy if the ER doc knows absolutely nothing about software development other than if he gets the decision wrong his life will be ruined.

As a somewhat more neutral analogy, waving prices for github vs VSS in front of my mother quite frankly holds negative value for all participants, its a Potemkin show that merely wastes labor hours. Pretending to provide a market where none can possibly exist is wasting labor time putting on a show. My mom has some skill in real estate law so she should be making decisions in her area of expertise, and software professionals should be making decisions in their areas of expertise. My mom does not walk up to random people on the street and insist they make life changing decisions in minutes about the right legal strategy to provide clear title to real estate or WTF exactly she did before she retired, "well in that county a title search costs $X but title insurance and hoping for the best costs $Y unless we take it to court then it costs $Z..." etc.

See you can "make a market" in a campy going-thru-the-motions sense by having a semi-monopoly provider declaring an arbitrary list of prices on a chalkboard then calling that a free market. But making a functioning efficient market is a whole nother kettle of fish that requires a calm rational meeting of equal minds at a similar level of training and education to dispassionately trade interchangeable identical commodities, with the goal of all participants minimizing the spread between price and value. That's pretty much the opposite of good medical care.

I'm not really sure what your point is, so perhaps I missed something?

Conflating a want for price transparency, so that I can spend rationally where that rationality can be logically exerted doesn't really seem to me a free market ideal as much as it is just common sense. There are many medical decisions made that aren't life or death, and in my admittedly anecdotal experience, that's been most of them. A great many medical demands are inelastic.

When I lopped the (very) tip of my pinky finger off in an accident, it was nowhere near a dire enough injury to worry about the cost of repairs, but it did need stitches. Had there been a repair option, I might have opted for it, but only if the cost was reasonable.

That said, we already know the market value of an ER doctor, or a radiologist; that's how we know what to pay them. It's not as though they're members of some secret cult of faceless men exacting arbitrary payments for their efforts. We know what it takes to pay them, and presumably, they could come up with an hourly cost for their labor plus materials if they so chose. It might take some practice, as estimation is hard, but basically every other service industry in the world is able to do it, so I can't imagine why the medical industry would be exempt.

Regardless, sure, there are plenty of ailments that would hard to estimate, or meaningless to do so, but I don't see why that means nobody should try to do their jobs in a cost effective manner.

Indeed, hindsight does tend to clarify things significantly.

Frankly, don't you think the better solution might have been to cue in on your doctor's hint that the additional testing wasn't needed?

Maybe costs would have clarified things. Maybe cost isn't a meaningful proxy for medical judgment.

It's been some time, but if I'm recalling correctly, the doctor seemed in favor of the additional testing. Whether that was because he was genuinely concerned, scared of a malpractice suit from insufficient testing, or something else I'm not privy to, I wouldn't know.

As I mentioned elsewhere, maybe cost isn't a meaningful proxy for medical judgement, but on relatively trivial matters like wrist sprains, minor cuts, etc., it may often be a good way to discriminate, assuming appropriate follow-up to ensure that the initial decision wasn't errantly done.

I don't understand how paying directly or not has anything to do with choosing a surgeon. Plently of socialized medicine system you get zero choice. Some 1/2 socialized, you still have to pay if you want a choice in surgeon.

I'm not against socialized medicine I asking how do you think not having "market forces" fixes the problem you're suggesting

One problem is that medicine is very counter-intuitive in terms of evaluating practitioners. Metrics that consumers think they would want to look at are very deceiving once you start digging into what's really going on. The reality is that often the best doctors work on extremely challenging cases that have both high costs and poor outcomes. Some doctors build practices that filter to select the easy cases (i.e. those with a large mismatch between allowed charges and actual resource needs). This allows them to push discounts do their patients aggressively because of the overall pool of cases they've selected. I haven't seen of any reasonable way of handicapping metrics that actually works, and frankly I would be extremely wary of anyone that claimed to have the answer. Honestly, I have no idea how you go about evaluating this situation as a consumer.

There are some attempts to do severity adjustment, though these typically rely on diagnosis codes, which can themselves be gamed.

A fairly strong heuristic is simply experience: the number of procedures performed is highly correlated with quality. Facilities often have strongly uniform outcomes determined by senior staff.

"The problem with asking patients to shop based on price is that the penalty for making the wrong choice is much more devastating than with just about any other good or service."

Yet we do it with our sight when choosing laser surgery. I get the feeling that better institutions would win out and care would improve.

I think preventative care is a separate issue. The government, instead of passing that horrible bill (and yes, I read every page and cussed at every will be decided later line) could have just subsidized a yearly examination, 2 teeth cleanings, expanded health saving accounts, and attached a "disaster clause" to all insurance where the government picks up any cost for an incident over $100,000 (changes the cost of insurance by providing a cap).

Laser eye surgery is a completely elective procedure. It can't be compared at all to the non-elective procedures that contribute most to healthcare costs.

You need to correct your vision in some way (glasses, contacts, surgery). It is the surgical variant of the correction. It demonstrates how the market for surgical procedures works when considering pricing. Its all markets, its just who gets to decide on what you value.

Consumers have a lot of "skin in the game" right now between being uninsured, or having significant co-insurance, or high-deductible/catastrophic plans, etc., but they do not generally have the information to make informed decisions about their own care. Some research suggests patients with these kind of plans do spend less, but largely by foregoing preventative care: http://commonhealth.wbur.org/2012/05/consumers-health-costs

As for the cost-benefit thing, do you think insurers are not doing it?

Two interesting problems.

One is that a free market might result in "better" purchaser decisions, but a free market requires a meeting of the minds and some patients won't have minds, temporarily or permanently, and those that do have a staggering power imbalance of one uneducated untrained individual vs a thousand person highly trained organization. We're not talking about two commodity traders exchanging with each other. This is before we get into "state of mind" and panic, someone having a heart attack can't focus on contractual negotiations quite like the guy across the table not currently having a heart attack. And then there's the timeliness issue, and geography...

The second, is medical care cannot be treated as a commodity. I was horrified at the "taylorism gone wild" in the article where one surgeon used twice the bandages as the other, and therefore is bad. In the real world that will be implemented as "well, yes I know your tumor was a completely different size and shape than his tumor, and bandages cost almost nothing compared to a human life, but we have beans to count and quotas to stay under so you'll just have to bleed to death because it would take too many bandages to save your life" or "well yes half his patients are permanently crippled and none of hers do, but she uses twice as many band aids, so we'll have to downsize her and keep him, to stay under our bandage budget". Without a commodity of interchangeable identical product, you can't have a free market or at least not the gains of a free market. Its interesting that the medical system admits as such... there are specializations of patient disease clusters, but despite the best efforts of centuries of work we still can't specialize beyond a very shallow level, we have cardiologist specialists not specialists in 25% arterial blockages of the left ventricle, even though its "obvious" that if it were implementable, specialization to that level would be very helpful. This is before we get to the issue that you get what you count, so if you count meaningless metrics you'll get excellent meaningless metrics, not health care.

A better model for the economics of health care would be fine art. Or maybe education. Notice how the money spent correlates strongly with how much money is available, and has little correlation with results, at any level beyond bare requirements. Or how attempts at taylorism style management provide laughable comedy results. Health care does seem to "fit" better with fine art or education, than with accounting or a manual factory assembly line.

> I was horrified at the "taylorism gone wild" in the article where one surgeon used twice the bandages as the other, and therefore is bad.

It didn't say that it was bad; it just noted it as a fact. Perhaps Dr. Jones's patients need more bandages because of the types of patient he sees (which might indicate that he's really good at what he does); perhaps his incisions are larger (which would be bad); perhaps he just wastes bandages (which is also bad). All measurements can do is demonstrate a difference: good management is required to discern the reason for the disparity, and determine if it's a positive or negative indicator.

Its possible the reason is unknowable at this time. As an example one theoretical reason you didn't mention is changing bandages twice as often reduces post-op infections by 75%, although no one has discovered or proven that yet, its sounds possible, and post-op infections are expensive and painful compared to cheap bandage replacement. Minimization of cost of any random individual component almost always doesn't result in minimization of total system cost.

Its important to build robust fault tolerant fail safe systems. By definition, half of all management will be below median. A system designed to give the bad ones maximally overpowered weapons to wreck havoc is not a good system design. Numerology as a weapon can be destructive.

It seems reasonable that as a system it might enable the top half of managers to take three steps forward, while enabling the bottom half of managers to take five steps back, not forgetting the fixed cost due to documentation and processing labor (bean counting) and micromanagement somewhere around a fixed two to four steps back just to gather and process the data, regardless if its used or not. That's a badly designed system.

This is before design criticism along the lines of never implement a system more complicated than its minimal need. Simplicate and add lightness. The substantial money spent counting, processing, and graphing every band aid could have been spent more simply, probably more productively, by spending $X extra on surgeon hiring salary to get a slightly measurably higher class of surgeon... or manager... Or the mental effort spinning around inside a surgeon's brain while cutting could be spent on improving patient outcomes by perfecting incisions or sutures or during surgery judgement calls ... or those brain cycles can be spent on minimization of band aids to meet band aid quota.

It seems so right from an Adam Smith invisible hand point of view: offer choice, allow people to shop around, and costs will go down, right?

The US has already been running this experiment—the results are in, and it's a huge loser. Everything you need to know is in this chart:

https://en.wikipedia.org/wiki/File:Total_health_expenditure_...

It turns out that everyone pays a lot less when you treat healthcare infrastructure like roads, dams, the power grid, or any other infrastructure— it has certain high fixed costs which build and sustain it (financed from the general tax base), and apart from that, it should be available for use by everyone.

What we have isn't really insurance, but health care supplements. Insurance is for unforeseen events. But US health insurance pays for regular checkups and procedures.

And I'll take our more expensive health care over any other system any day of the week. Because when the day comes when 1 of the 2 primary killers in the US comes for me (cancer and heart disease), no country in the world has better outcomes. And the US does it for a much more diverse, much larger, and unhealthy population. Cancer and heart disease outcomes are the best proxy for how well a health care system operates.

> Cancer and heart disease outcomes are the best proxy for how well a health care system operates.

Maybe for you. There are plenty of other fundamental measures of health care quality, like infant mortality rate.

Wouldn't infant mortality rates have a lot more to do with prenatal care in industrialized nations? I doubt more kids in the US are dying because of a lower standard of care and more likely because of more drug addicted/alcoholic/poorer nourished moms. Using infant mortality as a proxy in an industrialized nation is just hacking the stats.

If you don't think such things matter, then explain how Hispanic women on the west coast have both the lowest access to health care and also the lowest death rate from cancer? DNA? Diet? Low stress? Could be. But either way, their exceptionally low cancer rates don't reflect upon the US health care system.

What if you moved somewhere with lower cancer rates in the first place?

So I'd have less chance of getting cancer because I smoked less and rode bikes more, but if I did get cancer (and in the end nearly everyone in industrialized world dies of heart disease or cancer), I'll have a significantly lower chance of surviving?

And what about breast cancer? The US kicks major arse in this regard, and breast cancer rates are pretty steady across countries.

>> and breast cancer rates are pretty steady across countries

No, they are not. See Japan. Eat your Kelp. http://www.berkeley.edu/news/media/releases/2005/02/02_kelp....

I find that one interesting because they suggest not to eat a lot of kelp in part due to the high iodine content. But lots of others (self included) believe the iodine is the key.

Do you have a source for US outcomes being better? http://www.washingtonpost.com/news/to-your-health/wp/2014/06... http://theincidentaleconomist.com/wordpress/the-state-of-us-...

That was an interesting read, lots of good points. But I think that articles points are orthogonal to mine. I agree the US is a much less healthy place than many other industrialized countries. But that isn't a reflection on our health care industry, that's a cultural issue. If the US has a higher rate of breast cancer remission, but an even higher rate of breast cancer occurrence, that would reflect well on the health care system while resulting in a increased years of life lost.

You're doctor can't make you stop being fat, he can't make you jog, and he can't make you stop stressing about not jogging and being fat :P But in the US he can give you an edge over other countries in beating cancer.

It's an extremely tough problem to determine what is cultural causes, what is genetic, and what is health care. But just normalizing on age incidences, the US does well. I bet if we adjusted for relative health (impossible, but I'm wishing for perfect data), the US would do remarkably better.

If I have time later today, I'll dig up my sources. But I lost interest in this years ago when people overwhelmingly thought that because the US has a higher mortality rate, our health care system must suck :/

I'll take our more expensive health care over any other system any day of the week.

You pay almost twice as much per capita compared with Canada. Does it really all go toward better general care and longer lifespans? I don't have link handy, but my understanding is that a huge amount of it goes to the bureaucracy associated with competitive for-profit insurance (including bailouts and so-on), not to mention crisis treatment for those who have avoided proactive care for fear of a big bill or being prescribed unnecessary and expensive drugs.

I'm not necessary dismissing your use of cancer and heart disease as proxies for care quality (dubious though that is), more just asking if you really think that that small gain is where the gigantic premium in US health care cost goes.

> And the US does it for a much more diverse, much larger, and unhealthy population. Cancer and heart disease outcomes are the best proxy for how well a health care system operates.

Since the purpose of a health care system is to maintain the health of the population, overall population health is a much better thing to look to that micro outcomes from two particular events.

And by overall health, as you yourself note, the US doesn't do particularly well compared to other developed countries, despite spending far more as a share of GDP than most (and much more per capita.)

Your claim is a lot like claiming that an organization has a better software quality system than others because it fixes bugs in production slightly faster than others, even though it also allows far more bugs into production than the others.

>offer choice, allow people to shop around, and costs will go down, right?

Just wanted to point out that tough medical licensing regimes actually make it quite hard to shop around. This paper provides an interesting analysis: http://econjwatch.org/file_download/54/2004-08-svorny-reach_.... Or, for the short-of-time, a shorter, more biased summary: http://www.forbes.com/2009/08/25/american-medical-associatio...

> It seems so right from an Adam Smith invisible hand point of view: offer choice, allow people to shop around, and costs will go down, right?

> The US has already been running this experiment—the results are in, and it's a huge loser.

No, we've really not been running that experiment: we don't offer much choice, we don't really allow people to shop around, and most importantly, the vast majority of people aren't paying directly for their own care.

The U.S. most definitely does not have a free market in health care, and hasn't for decades.

I agree with the other comments: we don't have a free market now and health coverage comes with huge amounts of moral hazard for both providers and patients.

Veterinary medicine is something operates in a truly free market environment, and it seems to work towards efficient outcomes, so I'm curious if the same could be done for humans.

Well, would you be OK with people saying "this procedure is too expensive so we're just going to put Grandma down?" I don't think veterinary medicine is really quite the same.

One of the difficulties of improving the overall healthcare system is there are very few ways of making systematic improvements. This article talks about a single hospital. They demonstrated a method which works, but this only improves one hospital. For a real change to occur this would need to be repeated at a large number of hospitals on a hospital by hospital basis. The US healthcare system is incredibly fractured with large regional variances. The question is how to improve the system as a whole not on a piece by piece basis.

The journey of a thousand miles begins with one step.

-- Lao Tzu

Reply on news.ycombinator.com